|
[Previous entry: "How to sit at a cafAMPeacute;"] [Main Index] [Next entry: "The best of Blogathon 2003"]
"Six degrees of separation or unification?"
08/01/2003 Entry
Every Internet user knows that the network helps us easily connect with numerous people across wide distances. Over the years, various services have sprung up to facilitate such connections. But is it always a good idea to join such networks? Is bigger always better?
There has been an increasing amount of discussion recently about Friendster, a relatively new online service based on connecting people to others through their existing social networks. The service has been enthusiastically covered by the likes of The Village Voice, Wired and Slate describing how Friendster is the newest great social - and procrastinatory - tool given all the information one can glean about people in one's social circles through the service.
Since newspapers specialize in covering the news not the olds perhaps it is not surprising that they will always want to find a new twist on something even if it is simply the repetition of something old. But Web sites like this existed in the past and had trouble surviving. SixDegrees and PlanetAll were two extremely similar services in the 1990s, but both were victims of the Internet bust and ceased operations years ago despite the fact that they had as many users as Friendster does today.
Many express much excitement about Friendster and even claim that it is the next Google in online popularity and importance. I do not share this enthusiasm and here is why.
People tend to focus on the seemingly fun and useful aspects of Friendster. However, it can have serious shortcomings. Beyond the more obvious concerns regarding the truthfulness of information conveyed on the sites (the most extreme but quite transparent version of which are such bogus accounts as Jesus and God that people have created), networks like this can dilute the importance of social ties and can lead to uncomfortable situations when acquaintances from different parts of one's social circles intermingle.
It may be a good idea to take care in signing up to such a service and indiscriminately adding "friends" because it is not always ideal for different networks of one's life to intertwine too much. We are all members of various social groups. We have family ties, workplace ties, friends from childhood, friends from high school, from college, possibly graduate school, not to mention our myriad of other social relations through other affiliations. There may be reasons people don't want to see these different worlds overlap too much.
A fairly obvious case is a person who is not very open about a particular aspect of her identity (e.g. sexuality, religious affiliation, political convictions) in one realm of her life, say at work, but has a group of friends who are very clear about that part of her identity. Does she necessarily want these different worlds intertwining endlessly? Even if someone does not have any major hidden identities, we all play different social roles depending on the situation and people around us so things can get complicated if these criss-cross too much without boundaries. Even if you think you are one of those people who has nothing to hide and who behaves exactly the same everywhere, just think about interactions between you and your parents, you and your boss, you and your closest friends, these are not fully interchangeable no matter who you are and how you live your life. (See Irving Goffman's Presentation of Self in Everyday Life and George Herbert Mead's Mind, Self and Society for some basics on this.)
Another problem with the service is that it is unclear what criteria people use to approve others as "friends". I suspect many people approve most of those who get in touch with them because it would be rude not to do so. Moreover, in so far as people are trying to build a bigger network - which is probably not the case for all users, but likely for many - it is to their advantage to create additional "friendship" ties. But if these links are not very strong then a second or third-level tie will be extremely weak and it is not clear how beneficial it would be to anyone. How much can we trust these ties? An important part of contacting people via acquaintances we share is a level of built-in trust that comes from mutually knowing a third party. But if users link to people they barely know and those people link to others they barely know then the ties are going to get extremely weak and little trust will flow through them wiping away one of the core ideas behind the entire enterprise.
A bigger network is not always a better network. If the criteria for creating ties between members is too unclear or becomes too diluted (what exactly constitutes a "friend" on Friendster?) then the ties become close to meaningless. And although you can still map someone's path to you, there is no longer much reason to trust them or feel much affiliation to them any more than the person sitting next to you in a cafe.
The apparent rush to sign up on Friendster reminds me of the enthusiasm with which people started participating in various online discussions in the 1990s. People would post all sorts of information and comments not realizing that these would be available to large audiences. Along came better search engine indexing and now these less-than-flattering comments made about college flings and annoying coworkers are available to colleagues and supervisors at the click of a button. Similarly, it seems, people are flocking to Friendster to share all sorts of information with networks larger than they can imagine only to realize in time - or so I suspect - that they shared more than they may have wanted and madetthemselves available to more people than necessarily ideal. I realize it is probably possible to delete an account, but that often only occurs to people once some damage has been done.
I do not mean to sound paranoid and I do not mean to suggest that everyone has all sorts of ghosts in their closet. It is simply that everyone has different social roles and mixing these too much indiscriminately may lead to some complications.
Often shared affiliation with a group or organization will be more helpful than being able to link to someone through even just two or three degrees of connections. Anyone who has relied on alum ties can attest to this. The movie Legally Blonde 2 presents a good example of this. The central character is able to win the attention and support of an important political actor thanks to their sorority ties identified via the rings they both wear.
Some schools have leveraged the importance of such ties quite well through creating communities based on their alums. People in those groups may not know each other but they share a group affiliation that may be important to many of them. And although some may claim that such ties are elitist or irrelevant because simply sharing an institutional affiliation won't make you best buddies with someone (true), it would be hard to argue that even two or three degrees of separation from a person via very tangential "friendship" ties has much larger chances of mutual understanding and support.
Maybe they call it "six degrees of separation" for a reason.
Replies: 6 Comments have been posted, click here to see them and add your own
Interesting.
I just signed up for Friendster -- well, had myself invited, as it were -- to see what it was all about. Not sure what I think about it yet, because haven't used it much yet.
So intuitively, at least, I kind of agree with both your criticisms, Eszter. But both sound awfully like constructive criticism which could well be taken on-board and slowly incorporated into the design.
1) On the friend/not-friend binarism: the problem to solve becomes moving from a simple opposition to a more complicated spectrum. Or from digital to analogue. Or from discrete to continuous variables.
The latter being the point, I think. Solving it probably isn't a matter of thinking up a growing nunmber of boxes in how to define relationships between people ("friend" and "colleague" and "acquaintance", etc).
Rather, they'd want to develop some sort of algorithm for measuring the intensity of each dyadic relationship based on interaction between the two user identities. So that you'd authorise an "acquaintance" or a "hello", and any more closeness than that would derive from the quality and frequency of interaction, according to the really clever algorithms that the vast team of sociologists had designed.
'Course, the problem is that I don't really know how much user interaction is built into Friendster vs how much assumed to happen offline. If everything happens offline, then the problem is far harder to solve: call it an inherent flaw in social relationship software that lives on a separate plane from the social relationships themselves.
2) On maintaining distinct social networks: should be do-able in several ways.
The really fancy version builds on (1) and has something to do with measuring how well-connected the given networks are -- thus mapping out separate social formations -- and then deal with them in its configuration. Something to do with only seeing network members where the network meets a certain connectivity score, etc.
The not-so-fancy version has one registering several accounts, and being rather careful not to let them bleed into each other. Which is rather how we negotiate separate social networks in meatspace...
Posted by Bram Abramson @ 08/01/2003 05:00 PM CST
Well, strength alone isn't the issue when talking about ties - there are other metrics. What about people who are your friends because of shared interests but you consider completely financially irresponsible? What about people who are your good friends but you wouldn't want your parents to know you know?
Posted by David Brake @ 08/02/2003 07:24 AM CST
Wow -- back again!
Well, strength alone isn't the issue when talking about ties - there are other metrics. What about people who are your friends because of shared interests but you consider completely financially irresponsible?
This speaks to the issue of social software as representation of social experience. Like any map or statistical data set, attempts at perfectly mimetic representation are a bit beside the point, I think, and have to do with losing sight of one's goals. The famous Borges quote holds re maps as big as the things they represent.
So the trick is not to get caught up in a giant intellectual exercise to rebuild the world in a machine. Rather it's the uses of, here, the software that become the basic question: what do I want it to be good for? What might work well? What do I need to express in order to make that happen?
Relationship intensity is the gloss on a basic feature behind Friendster. The challenge of expressing other vectors of friendship in social software is interesting, but the more creative part is figuring out which vectors one wants to. Perceived financial responsibility might be useful for an online microlending network, or -- same thing -- if one wanted to add online microlending features to Friendster. For Friendster itself, perhaps there are indeed more specific aspects to relationships that it would be both workable and useful to model. And so forth; that's one of the things research is supposed to do, methinks.
What about people who are your good friends but you wouldn't want your parents to know you know?
Exactly ... see (2) above.
Posted by Bram @ 08/02/2003 12:19 PM CST
I don't quite understand the problem and maybe I am misreading the argument, but it sounds awefully like - oh no, people might get themselves into a serious bind here and not know it! Friendster is a way to connect, its fun, its wildly popular, and... forgive me for the following heresy, but I think its mostly useless.
Its easier to send messages to friends via email, friendster's interface is slow, often inaccessible and generally clunky. So if you need to talk to friends you email them. So why is this thing so popular? Why do teenagers collect people in their cell phone addressbooks and show them off? Why are buddy lists for IM users so long when they usually chat with only a fraction of the people on them? People like to feel connected. They like to see that they are well... popular, they like to show it off. Friendster is an easy tool for this. A way to validate self in a digital world as someone about whom people care. It has other side effects - people do connect and find each other on these networks (like any other networks, news groups, etc...) Friendster is a social space and as such requires a set of social conventions that people use to deal with it.
I certainly don't think there is reason to be concerned with this "mixing of identities and degrees of separation" though. People generally know how to manage this themselves. Just because this is something of a new idea, doesn't mean we suddenly forget how to function in its social sphere.
Posted by Irina @ 08/06/2003 05:33 PM CST
First of all, I'm writing from Budapest, and I'm always thrilled to run across Hungarians abroad, especially postgrads. (That seems like a Friendster kind of thing to say.) I'm also glad this blog is here, and that people are thinking this way about Friendster.
I've also been thinking about this "mixing of identities" issue. I'd like to do a little piece about it, but I'm not sure how yet, or why. One of the things we should be aware of is that friendster is highlighting for us something that exists in real, physical society. As you say the issue is that people already participate in a wide variety of different social groups, discourses, etc., in their real lives, and these discourses can be vastly diverse and conflicting. Like you say, we have Goffman, Bakhtin, a bevy of sociolinguists, Joyce, and a lot of others to open our eyes to different aspects of this fact.
In the real world, though, people are limited to having only one body, and in most cases, only one or two telephone numbers, addresses, etc. Email addresses are, of course, different, and on Friendster, they can create as many profiles as they want. And they do.
Take the example of pornstar Aria Giovanni - for whom there are two profiles on Friendster under her stage name. One or both of these profiles may have been set up by fans - or by Aria herself. Under her personal name she may manage another circle or aspect of her personality. Or, she may not even know that Friendster exists. We might say that a part of this real person's social identity is something almost like a semifictional character, like the Muppet "Beaker," or Pippi Longstocking. (Friendster has at least eight registered Pippi Longstockings.)
If we look only at what's on the screen, Friendster probably won't give us at all an accurate picture of real life. But if we ask her about it, we might find that's a part of real strategies for managing multiple social networks and spheres. So the Friendster "fiction" here is still a part of something which is definitely social reality, and far from being a threat to the validity of the whole system, it might even represent a native attempt to address these issues you've picked out.
Friendster's very liberal policies compared to, say, www.hotornot.com about fictional personalities, copyright violations with pictures, etc. probably help with that, and also distinguish it from www.sixdegrees.com, www.wiw.hu, and others.
Also, on Friendster we get abstractions like "user" profiles that refer to institutions, famous authors, or even ideas, social statuses, etc., which people use as a tool to express institutional affiliations, etc.
Some cases in point are: profiles for countries, universities, cities; profiles for social statuses like "Bachelor's Degree;" even professional clusters like "Casting Corner."
Maybe in these cases, we have to get beyond a literalistic view of Friendster, like the one Friendster gives of itself in its "join" pages, in order to understand how the site is working to let people network or model their social identities.
So, my answer to this problem of not being able to be sure how "real" people's social connections are is that, depending on how you define the problem, it may not be a meaningful issue.
Of course, actual social relations are much more complex than what this software represents, and so are personal identities. I guess rather than considering how successful the software is at modelling reality, I might be more interested in considering how Friendster is working as a social phenomenon in itself. As far as negotiating social spheres in netspace, I like the way Friendster lets people have several profiles, fictionalize themselves, or lie. Of course that makes it less useful as a tool for microlending - for that we have credit ratings, and other real indexes of people. Friendster is something completely different, I'd say, itself a social sphere with a developing discourse and set of conventions of its own.
Any comments to me by email please! adam@liberte.li
Posted by Adam Warner @ 08/16/2003 06:47 PM CST
For a map rather than a service check out http://www.knowmates.com
Posted by Jamie Morrison @ 09/25/2003 09:22 AM CST
| 
